CN111275090B - GNSS ultra-fast clock error forecasting method - Google Patents
GNSS ultra-fast clock error forecasting method Download PDFInfo
- Publication number
- CN111275090B CN111275090B CN202010049690.5A CN202010049690A CN111275090B CN 111275090 B CN111275090 B CN 111275090B CN 202010049690 A CN202010049690 A CN 202010049690A CN 111275090 B CN111275090 B CN 111275090B
- Authority
- CN
- China
- Prior art keywords
- clock
- value
- forecasting
- clock error
- adopting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
Abstract
The invention provides a GNSS ultra-fast clock error forecasting method, which improves the current situations of low clock error real-time forecasting precision and poor stability of the existing GNSS clock error forecasting method, and comprises the following steps: preprocessing the clock difference data; performing principal component analysis on the clock difference data; modeling and forecasting the main components and the total residual sequence respectively; obtaining a final forecast value; the invention can be widely applied to the technical field of satellite navigation.
Description
Technical Field
The invention relates to the technical field of satellite navigation, in particular to a GNSS ultra-fast clock error forecasting method.
Background
In the real-time navigation and positioning of a Global Navigation Satellite System (GNSS), the precision of a satellite clock error product can directly influence the service capability of high-precision navigation and positioning time service, and in order to further improve the precision of clock error prediction and improve the current situation that the current clock error real-time prediction precision is low, scholars at home and abroad make a large number of researches on prediction methods; the influence of the model index coefficient on the grey model prediction precision is large; although the spectrum analysis model considers the period term in the clock difference, the period in the clock difference can be more accurately determined only by a longer clock difference sequence, and the advantages of the model can be exerted only by modeling longer clock difference data during fitting and forecasting; for the wavelet neural network model, the determination of the network topology structure has difficulty; for an abnormal complex non-stationary and non-linear random sequence of satellite clock error, a single model is difficult to accurately express and effectively forecast, although the combined model can consider more influences of random terms on the forecast than the single model, most combined models are simple combinations, the combination is not carried out according to the characteristics of the single models, the advantages of the combined model are not well played, and the forecast precision and the forecast stability have larger promotion spaces.
Therefore, most of the existing models do not consider the influence of randomness in clock error characteristics and system noise errors on clock error prediction model modeling, which is one of the reasons that the clock error real-time prediction precision and stability of most of the current clock error prediction models are low, and the clock error real-time prediction precision and stability can be further improved.
Disclosure of Invention
The invention provides a GNSS ultra-fast clock error forecasting method with high forecasting precision and good stability, aiming at the technical problems of low real-time forecasting precision and poor stability of the existing GNSS clock error forecasting method.
Therefore, the GNSS ultrafast clock error forecasting method provided by the invention is realized by the following steps:
step 1: preprocessing the clock difference data;
step 2: performing principal component analysis on the clock difference data;
and step 3: modeling and forecasting the main components and the total residual sequence respectively;
and 4, step 4: and obtaining a final predicted value.
Preferably, the clock difference data is preprocessed: and (3) after the clock error data are converted into frequency data, removing gross errors by adopting a median method, and completing by adopting a linear interpolation method.
Preferably, the clock difference data is preprocessed: and predicting by adopting a polynomial model and setting a threshold value to judge whether clock jitter exists in the clock skew data, and if the clock jitter exists, performing sectional processing on the clock skew data.
Preferably, the principal component analysis of the clock error data: the clock error data mainly comprises a trend term, a period term and noise, the clock error is decomposed by adopting principal component analysis, most of noise terms are separated, only the trend term and the period term in the clock error are left, the trend term and the period term are used as principal components, and the noise is used as a secondary component A.
Preferably, modeling prediction is respectively carried out on the main component and the total residual sequence: modeling and forecasting the main components by adopting a robust spectrum analysis model to obtain a forecast value C, obtaining a fitting residual B of the main components, wherein the fitting residual has a certain influence on clock difference forecasting, adding the fitting residual of the secondary components A obtained after clock difference solution to form a new residual sequence A + B, and then modeling and forecasting by adopting a machine learning algorithm to obtain a forecast value D.
Preferably, the final predicted value is obtained: after the two prediction values C and D are added to obtain a new prediction sequence, the prediction sequence is integrally translated by utilizing a quadratic polynomial model and the difference value between the initial value of the last four epoch predictions of the clock difference and the initial value in the prediction sequence C + D to obtain a prediction value E; and (3) obtaining a new slope value by adopting a quadratic polynomial model and the last four epochs of the clock difference data, further obtaining a weighted average value of the new slope value and the slope value obtained by integral fitting, and performing slope deviation correction on the obtained forecast sequence E by utilizing the difference value of the new slope weighted average value and the slope value obtained by integral fitting to obtain a final forecast value F.
The GNSS ultrafast clock error forecasting method not only considers the randomness error, but also weakens the influence of the randomness error on modeling, delays the accumulation of the forecasting error by performing starting point deviation correction and slope deviation correction on a forecasting sequence, adopts ultrafast and precise clock error products of various navigation positioning systems to perform experiments, greatly improves the forecasting precision, improves the stability to a certain degree, and can control the influence of abnormal errors or clock error data with large partial deviation after data preprocessing on the forecasting precision.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
As shown in fig. 1, the present invention provides a GNSS ultrafast clock error forecasting method, which is implemented by the following steps:
step 1: preprocessing clock difference data
The method comprises the steps that due to the influence of an external environment, clock error data inevitably have gross errors, the existence of the gross errors can seriously influence the forecasting precision, so the gross errors are eliminated, the clock error data are converted into frequency data, then the gross errors are eliminated by adopting a median method and are supplemented by adopting a linear interpolation method, in addition, polynomial model forecasting is adopted, a threshold value is set to judge whether clock jitter exists in the clock error data, and if the clock jitter exists, the clock error data are processed in a segmented mode.
Step 2: principal component analysis of clock difference data
Because the clock error data mainly comprises a trend term, a period term and noise, the noise has certain influence on modeling when the spectral analysis model is used for modeling, and in order to weaken the influence of the noise on the clock error modeling, the clock error is considered to be decomposed by adopting principal component analysis, most of noise terms are separated, only the trend term and the period term in the clock error are almost left, the trend term and the period term are used as principal components, and the noise is used as a secondary component A.
And step 3: modeling and forecasting main component and total residual sequence respectively
Modeling and forecasting the main component by adopting a robust spectrum analysis model to obtain a forecast value C, and simultaneously obtaining a fitting residual B of the main component, wherein the fitting residual has a certain influence on clock error forecasting, adding the minor component A obtained after decomposition in the step 2 and the fitting residual B to form a new residual sequence A + B, and then modeling and forecasting by adopting a machine learning algorithm to obtain a forecast value D.
And 4, step 4: obtaining the final predicted value
Adding the two forecast values C and D obtained in the step 3 to obtain a new forecast sequence, and performing integral translation on the forecast sequence by using a quadratic polynomial model and a difference value between an initial value forecast by the last four epochs of the clock difference and an initial value in the forecast sequence C + D to obtain a forecast value E; and (3) obtaining a new slope value by adopting a quadratic polynomial model and the last four epochs of the clock difference data, further obtaining a weighted average value of the new slope value and the slope value obtained by integral fitting, and performing slope deviation correction on the obtained forecast sequence E by utilizing the difference value of the new slope weighted average value and the slope value obtained by integral fitting to obtain a final forecast value F.
According to the technical scheme, principal component analysis, decomposition and prediction, an anti-difference spectrum analysis model, a machine learning algorithm, starting point deviation correction and slope deviation correction key methods are combined, and the final prediction effect is obviously improved.
However, the above description is only exemplary of the present invention, and the scope of the present invention should not be limited thereby, and the replacement of the equivalent components or the equivalent changes and modifications made according to the protection scope of the present invention should be covered by the claims of the present invention.
Claims (3)
1. A GNSS ultrafast clock error forecasting method is characterized by comprising the following steps:
step 1: preprocessing the clock difference data;
step 2: performing principal component analysis on the clock difference data;
the clock error data consists of a trend term, a period term and noise, the clock error is decomposed by adopting principal component analysis, most of noise terms are separated, only the trend term and the period term in the clock error are left, the trend term and the period term are used as principal components, and the noise is used as a secondary component A;
and step 3: modeling and forecasting the main components and the total residual sequence respectively;
modeling and forecasting main components by adopting a robust spectrum analysis model to obtain a forecast value C, and meanwhile obtaining a fitting residual B of the main components, wherein the fitting residual also has certain influence on clock difference forecasting, adding fitting residual A of the secondary components obtained after clock difference solution to form a new residual sequence A + B, and then modeling and forecasting by adopting a machine learning algorithm to obtain a forecast value D;
and 4, step 4: obtaining a final forecast value;
after the two prediction values C and D are added to obtain a new prediction sequence, the prediction sequence is integrally translated by utilizing a quadratic polynomial model and the difference value between the initial value of the last four epoch predictions of the clock difference and the initial value in the prediction sequence C + D to obtain a prediction value E; and (3) obtaining a new slope value by adopting a quadratic polynomial model and the last four epochs of the clock difference data, further obtaining a weighted average value of the new slope value and the slope value obtained by integral fitting, and performing slope deviation correction on the obtained forecast sequence E by utilizing the difference value of the new slope weighted average value and the slope value obtained by integral fitting to obtain a final forecast value F.
2. The GNSS ultrafast clock error forecasting method according to claim 1, wherein the step 1 is implemented by: and (3) after the clock error data are converted into frequency data, removing gross errors by adopting a median method, and completing by adopting a linear interpolation method.
3. The GNSS ultrafast clock error forecasting method according to claim 2, wherein the step 1 is implemented by: and predicting by adopting a polynomial model and setting a threshold value to judge whether clock jitter exists in the clock skew data, and if the clock jitter exists, performing sectional processing on the clock skew data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010049690.5A CN111275090B (en) | 2020-01-16 | 2020-01-16 | GNSS ultra-fast clock error forecasting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010049690.5A CN111275090B (en) | 2020-01-16 | 2020-01-16 | GNSS ultra-fast clock error forecasting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111275090A CN111275090A (en) | 2020-06-12 |
| CN111275090B true CN111275090B (en) | 2021-07-16 |
Family
ID=71003485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010049690.5A Active CN111275090B (en) | 2020-01-16 | 2020-01-16 | GNSS ultra-fast clock error forecasting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111275090B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113240076B (en) * | 2021-04-27 | 2023-11-28 | 北京无线电计量测试研究所 | Clock error abnormal data detection method based on clustering and neural network algorithm |
| CN113608248B (en) * | 2021-06-25 | 2023-06-13 | 北京建筑大学 | Beidou 5G fusion high-precision patrol personnel positioning method and related equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107132558A (en) * | 2017-06-13 | 2017-09-05 | 武汉大学 | The multi-frequency multi-mode GNSS cycle slip rehabilitation methods and system of inertia auxiliary |
| CN108196279A (en) * | 2017-12-23 | 2018-06-22 | 航天恒星科技有限公司 | Satellite clock correction based on real-time stream resolves and forecasting procedure |
| CN110376620A (en) * | 2019-07-24 | 2019-10-25 | 广州中海达卫星导航技术股份有限公司 | Real-time clock deviation forecasting procedure, device and computer equipment |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012151006A1 (en) * | 2011-03-22 | 2012-11-08 | Trimble Navigation Limited | Gnss signal processing with delta phase |
| CN102098119B (en) * | 2011-01-27 | 2013-09-25 | 电子科技大学 | Method for forecasting receiver clock error |
| US10488505B2 (en) * | 2014-05-30 | 2019-11-26 | The Boeing Company | Positioning in indoor locations and other GPS-denied environments |
| CN104392414A (en) * | 2014-11-04 | 2015-03-04 | 河海大学 | Establishing method of regional CORS coordinate time series noise model |
| US20170140645A1 (en) * | 2015-11-06 | 2017-05-18 | The Board Of Regents Of The University Of Oklahoma | Traffic monitoring system |
| CN108107455A (en) * | 2017-10-30 | 2018-06-01 | 千寻位置网络（浙江）有限公司 | A kind of satellite clock correction Real-time Forecasting Method based on phase hit |
| CN109001771B (en) * | 2018-06-04 | 2020-10-23 | 北京未来导航科技有限公司 | Navigation satellite and low-orbit satellite real-time clock error determining and forecasting method and system |
| CN110632625B (en) * | 2019-08-19 | 2022-04-26 | 中国矿业大学 | GNSS time series step detection and repair method |
-
2020
- 2020-01-16 CN CN202010049690.5A patent/CN111275090B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107132558A (en) * | 2017-06-13 | 2017-09-05 | 武汉大学 | The multi-frequency multi-mode GNSS cycle slip rehabilitation methods and system of inertia auxiliary |
| CN108196279A (en) * | 2017-12-23 | 2018-06-22 | 航天恒星科技有限公司 | Satellite clock correction based on real-time stream resolves and forecasting procedure |
| CN110376620A (en) * | 2019-07-24 | 2019-10-25 | 广州中海达卫星导航技术股份有限公司 | Real-time clock deviation forecasting procedure, device and computer equipment |
Non-Patent Citations (3)
| Title |
|---|
| Spatiotemporal Filtering and Noise Analysis for Regional GNSS Network in Antarctica Using Independent Component Analysis;Li W等;《Remote Sensing》;20190228;第11卷(第4期);第386页 * |
| 导航卫星快速定轨和预报方法研究;刘伟平;《中国优秀硕士学位论文全文数据库 工程科技II辑》;20120715(第7期);第C031-430页 * |
| 顾及周期项误差和起点偏差修正的北斗卫星钟差预报;艾青松等;《测绘学报》;20161215;第45卷(第S2期);第132-138页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111275090A (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111046564B (en) | Residual life prediction method for two-stage degraded product | |
| Ferreira et al. | On the block maxima method in extreme value theory: PWM estimators | |
| Öcal et al. | Business cycle non‐linearities in UK consumption and production | |
| Jeong et al. | 21 A perspective on application of bootstrap methods in econometrics | |
| CN102306302B (en) | Satellite clock error prediction method based on empirical mode decomposition (EMD) model and generalized autoregressive conditional heteroskedasticity (GARCH) model | |
| CN111275090B (en) | GNSS ultra-fast clock error forecasting method | |
| CN111582551B (en) | Wind power plant short-term wind speed prediction method and system and electronic equipment | |
| CN104166787A (en) | Aero-engine remaining life prediction method based on multi-stage information fusion | |
| CN110942194A (en) | Wind power prediction error interval evaluation method based on TCN | |
| CN110398368B (en) | FBM-based bearing inner ring fault residual life prediction method based on long correlation model | |
| Heng et al. | Probabilistic and deterministic wind speed forecasting based on non-parametric approaches and wind characteristics information | |
| CN109255728B (en) | Photovoltaic power generation power neural network prediction method based on chaotic phase space optimization reconstruction | |
| CN106056239B (en) | Product inventory prediction method and device | |
| CN116307215A (en) | Load prediction method, device, equipment and storage medium of power system | |
| CN111723982A (en) | Medium-and-long-term power load combined prediction method based on gray-Markov chain | |
| CN109816165B (en) | Wind power ultra-short term power prediction method and system | |
| Xing et al. | Comparative study of reformed neural network based short‐term wind power forecasting models | |
| CN110555784A (en) | wind power prediction method based on gated cyclic neural network wind speed correction | |
| CN112948462A (en) | Ultra-short-term wind speed prediction method based on improved singular spectrum analysis and Bp neural network | |
| Zhao et al. | On the accelerated scan finite Markov chain imbedding approach | |
| CN107563573B (en) | Solar power generation capacity prediction method based on self-adaptive learning hybrid model | |
| CN116191412A (en) | Power load prediction method | |
| CN106529805B (en) | Generator importance-based power generation system reliability evaluation method | |
| CN113917500A (en) | Satellite clock error forecasting method | |
| CN113821419A (en) | Cloud server aging prediction method based on SVR and Gaussian function |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 264200 No. 180, Wenhua West Road, Shandong, Weihai Applicant after: SHANDONG University Address before: 250061 Ji'nan City, Shandong Province, No. ten, No. 17923 Applicant before: SHANDONG University |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |